EP3616406B1 - Verfahren und vorrichtung zur bewegungsvektorbeschränkung für videocodierung - Google Patents

Verfahren und vorrichtung zur bewegungsvektorbeschränkung für videocodierung Download PDF

Info

Publication number
EP3616406B1
EP3616406B1 EP18801342.9A EP18801342A EP3616406B1 EP 3616406 B1 EP3616406 B1 EP 3616406B1 EP 18801342 A EP18801342 A EP 18801342A EP 3616406 B1 EP3616406 B1 EP 3616406B1
Authority
EP
European Patent Office
Prior art keywords
block
sub
derived
primary
mvs
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
EP18801342.9A
Other languages
English (en)
French (fr)
Other versions
EP3616406A1 (de
EP3616406A4 (de
Inventor
Tzu-Der Chuang
Ching-Yeh Chen
Yu-Wen Huang
Chih-Wei Hsu
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
HFI Innovation Inc
Original Assignee
HFI Innovation Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by HFI Innovation Inc filed Critical HFI Innovation Inc
Publication of EP3616406A1 publication Critical patent/EP3616406A1/de
Publication of EP3616406A4 publication Critical patent/EP3616406A4/de
Application granted granted Critical
Publication of EP3616406B1 publication Critical patent/EP3616406B1/de
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/50—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
    • H04N19/503—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving temporal prediction
    • H04N19/51—Motion estimation or motion compensation
    • H04N19/513—Processing of motion vectors
    • H04N19/517—Processing of motion vectors by encoding
    • H04N19/52—Processing of motion vectors by encoding by predictive encoding
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/50—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
    • H04N19/503—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving temporal prediction
    • H04N19/51—Motion estimation or motion compensation
    • H04N19/55—Motion estimation with spatial constraints, e.g. at image or region borders
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
    • H04N19/157—Assigned coding mode, i.e. the coding mode being predefined or preselected to be further used for selection of another element or parameter
    • H04N19/159—Prediction type, e.g. intra-frame, inter-frame or bidirectional frame prediction
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
    • H04N19/17—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object
    • H04N19/176—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a block, e.g. a macroblock
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/189—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the adaptation method, adaptation tool or adaptation type used for the adaptive coding
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/50—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
    • H04N19/503—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving temporal prediction
    • H04N19/51—Motion estimation or motion compensation
    • H04N19/513—Processing of motion vectors
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/50—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
    • H04N19/503—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving temporal prediction
    • H04N19/51—Motion estimation or motion compensation
    • H04N19/537—Motion estimation other than block-based
    • H04N19/54—Motion estimation other than block-based using feature points or meshes

Definitions

  • the present invention relates to video coding utilizing sub-block based coding tools for motion estimation/compensation.
  • the present invention relates to constraining motion vector associated with sub-block based coding tools in order to reduce required bandwidth.
  • PMVD Pattern-based MV derivation
  • Motion estimation/compensation is a powerful coding tool that has been used in various coding standards such as MPEG-2, H.264 and the emerging HEVC (High Efficiency Video Coding) standard.
  • the motion information derived at the encoder side has to be transmitted to the decoder side, which may consume sizeable bandwidth.
  • motion vector prediction (MVP) to coding a current motion vector (MV) predictively has been developed.
  • MVs motion vectors
  • MVP motion vector prediction
  • HEVC supports the Skip and Merge modes for MVP coding.
  • Skip and Merge modes a set of candidates are derived based on the motion information of spatially neighbouring blocks (spatial candidates) or a temporal co-located block (temporal candidate).
  • spatial candidates spatially neighbouring blocks
  • temporal co-located block temporary candidate.
  • the residual signal is forced to be zero and not coded. In other words, no information is signalled for the residuals.
  • Each merged PU reuses the MV, prediction direction, and reference picture index of the selected candidate.
  • up to four spatial MV candidates are derived from neighbouring blocks Ao, A 1 , B 0 and B 1
  • one temporal MV candidate is derived from bottom-right block, T BR or centre-block T CT as shown in Fig. 1 .
  • T BR is used first. If T BR is not available, T CT is used instead. Note that if any of the four spatial MV candidates is not available, the block B 2 is then used to derive MV candidate as a replacement.
  • removing redundancy (pruning) is applied to remove any redundant MV candidate.
  • the encoder selects one final candidate within the candidate set for Skip or Merge mode based on the rate-distortion optimization (RDO) decision, and transmits the index to the decoder.
  • RDO rate-distortion optimization
  • the "Merge” mode referred hereafter may correspond to “Merge” mode as well as “Skip” mode for convenience.
  • the MVP technique is also applied to code a motion vector predictively, which is referred as AMVP (Advanced Motion Vector Prediction).
  • AMVP Advanced Motion Vector Prediction
  • MVDs transmitted motion vector differences
  • MVPs Motion Vector Predictors
  • the AMVP scheme is used to select a motion vector predictor among an AMVP candidate set including two spatial MVPs and one temporal MVP. Therefore, an AMVP index for MVP and the corresponding MVDs need to be encoded and transmitted for an AMVP-coded block.
  • the Inter prediction direction to specify the prediction directions among bi-prediction and uni-prediction (i.e., list 0 (L0) and/or list 1 (L1)) associated with the reference frame index for each list should also be encoded and transmitted.
  • the left MVP is selected based on the first available one from Ao, A 1
  • the top MVP is the first available one from B 0 , B 1 , B 2
  • the temporal MVP is the first available one from T BR or T CT (T BR is used first, if T BR is not available, T CT is used instead).
  • T BR is used first, if T BR is not available, T CT is used instead.
  • the second top MVP can be derived if there is a scaled MVP among B 0 , B 1 , and B 2 .
  • the list size of MVPs of AMVP is 2 in HEVC. Therefore, after the derivation process of the two spatial MVPs and one temporal MVP, only the first two MVPs can be included in the MVP list. If after removing redundancy, the number of available MVPs is less than two, zero vector candidates are added to the candidates list.
  • Sub-PU temporal motion vector prediction Conventional Sub-PU temporal motion vector prediction (Sub-PU TMVP)
  • a Sub-PU Temporal Motion Vector Prediction (Sub-PU TMVP, also called advanced temporal motion vector prediction, ATMVP) mode is applied in the merge mode. That is, the Sub-PU TMVP is a merge candidate for merge mode. As shown in Fig. 2 unlike the traditional temporal candidate, the Sub-PU TMVP mode partitions the current PU into multiple Sub-PUs, and finds all corresponding temporal collocated motion vectors for each Sub-PU.
  • the current PU of size MxN has (M/P)x(N/Q) sub-PUs, each sub-PU is of size PxQ, which M is divisible by P, and N is divisible by Q.
  • an "initial motion vector”, denoted it as vec_init is determined for the sub-PU TMVP mode.
  • the vec_init can be the MV of the first available spatial neighbouring block of the current PU 210.
  • the MV of other neighbouring block may also be used as the initial motion vector.
  • LX L0 or L1
  • list L0 or L1
  • LX is better for collocated information
  • LX is equal to L0, and vice versa.
  • LX assignment can be performed at a slice level or picture level.
  • a “collocated picture searching process” then starts.
  • the “collocated picture searching process” is to find a main collocated picture for all sub-PUs in the Sub-PU TMVP mode.
  • the main collocated picture is denoted as main_colpic.
  • main_colpic Conventionally, it first searches the reference picture selected by the first available spatial neighbouring bloc. Then, in B-Slices, it searches all reference pictures of current picture starting from L0 (or L1), reference index 0, then index 1, then index 2, and so on (increasing index order). If it finishes searching L0 (or L1), then it searches another list.
  • P-slice it first searches the reference picture selected by the first available spatial neighbouring block. Then, it searches all reference pictures of current picture of the list starting from reference index 0, then index 1, then index 2, and so on (increasing index order).
  • the "availability checking” process checks the collocated sub-PU around centre position of the current PU pointed by vec_init_scaled, where vec_init_scaled is the MV with appropriate MV scaling from the vec_init.
  • Various ways can be used to determine the "around centre position".
  • the "around centre position” corresponds to the centre pixel. For example, if the PU size is M*N, the centre is equal to position (M/2, N/2).
  • the "around centre position" corresponds to the centre sub-PU's centre pixel.
  • the "around centre position” may be the mix of embodiment 1 or embodiment 2 depending on the current PU shape.
  • the detail implement of the "around centre position” is not limited to these 3 embodiments.
  • availability checking if the checking result is an inter mode, then the availability is true; otherwise (the checking result being an intra mode), then the availability is false.
  • current searched picture is labelled as the main collocated picture and the searching process finishes. If the availability is true, the MV of the "around centre position" is used and scaled for the current block to derive the "default MV". If the availability is false, then it goes to search next reference picture.
  • MV scaling is needed when the reference picture of the vec_init is not equal to the original reference picture.
  • the MV scaling process is that, a scaled version of the motion vector is used.
  • the MV is scaled based on the temporal distances between the current picture and the reference picture of the vec_init and the searched reference picture, respectively.
  • the scaled MV is denoted as vec_init_scaled.
  • Sub-PU_i_x means horizontal left-top location of sub-PU i inside the current picture (integer location)
  • Sub-PU_i_y means vertical left-top location of sub-PU i inside the current picture (integer location)
  • vec_init_scaled_i_x means horizontal part of vec_init_scaled_i, which has integer part and fractional part and we only use integer part in the calculation
  • vec_init_scaled_i_y means vertical part of vec_init_scaled_i, which has integer part and fractional part and we only use integer part in the calculation.
  • shift_x means a shift value. In one embodiment, shift_x can be half of sub-PU width but not limited to this embodiment.
  • shift_y means a shift value. In one embodiment, shift_y can be half of sub-PU height, but not limited to this embodiment.
  • Step 3 it finds the motion information temporal predictor for each Sub-PU, which is denoted as SubPU_MI_i, of each Sub-PU.
  • the SubPU_MI_i is the motion information (MI) from collocated_picture_i_L0 and collocated_picture i L1 on collocated location x, collocated location y.
  • MI is defined as the set of ⁇ MV_x, MV_y, reference lists, reference index, and other merge-mode-sensitive information, such as local illumination compensation flag ⁇ .
  • MV_x and MV_y may be scaled according to the temporal distance relation between collocated picture, current picture, and reference picture of the collocated MV. (Not limited to this embodiment). If MI is not available for some Sub-PU, then the MI of Sub-PU around centre position will be used (in another word, the default MV is used).
  • a spatial-temporal motion vector prediction (STMVP) is also included in merge mode coding.
  • STMVP the motion vectors of the sub-CUs are derived recursively following the raster scan order by using the temporal motion vector predictor and spatial neighbouring motion vector.
  • Fig. 3 illustrates the concept of STMVP.
  • the neighbouring NxN blocks in the current frame are labeled as a, b, c, and d.
  • the motion derivation for sub-CU A starts by identifying its two spatial neighbours.
  • the first neighbour is NxN block above sub-CU A (block c). If this block c is not available or is intra coded the other NxN blocks above sub-CU A are checked (from left to right, starting at block c).
  • the second neighbour is a block to the left of the sub-CU A (block b).
  • TMVP temporal motion vector predictor
  • a pattern-based MV derivation (PMVD) method is disclosed.
  • the pattern-based MV derivation (PMVD) method is also referred as FRUC (Frame Rate Up Conversion) in VCEG-AZ07.
  • the FRUC consists of bilateral matching for a bi-prediction block and temporal matching for a uni-prediction block.
  • Fig. 4 illustrates an example of FRUC (Frame Rate Up Conversion) bilateral matching mode, where the motion information for a current block 410 is derived based on two reference pictures.
  • the motion information of the current block is derived by finding the best match between two blocks (420 and 430) along the motion trajectory 440 of the current block in two different reference pictures (i.e., RefO and Ref1).
  • the motion vectors MV0 associated with RefO and MV1 associated with Ref1 pointing to the two reference blocks shall be proportional to the temporal distances, i.e., TD0 and TD1, between the current picture (i.e., Cur pic) and the two reference pictures RefO and Ref1.
  • Fig. 5 illustrates an example of template matching FRUC mode.
  • the neighbouring areas (520a and 520b) of the current block 510 in a current picture i.e., Cur pic
  • a template i.e., 530a and 530b
  • a reference picture i.e., RefO in Fig. 5
  • the best match between template 520a/520b and template 530a/530b will determine a decoder derived motion vector 540.
  • RefO is shown in Fig. 5
  • Ref1 can also be used as a reference picture.
  • a FRUC_mrg_flag is signalled when the merge_flag or skip_flag is true. If the FRUC_mrg_flag is 1, then FRUC_merge_mode is signalled to indicate whether the bilateral matching merge mode or template matching merge mode is selected. If the FRUC_mrg_flag is 0, it implies that regular merge mode is used and a merge index is signalled in this case.
  • the motion vector for a block may be predicted using motion vector prediction (MVP), where a candidate list is generated.
  • MVP motion vector prediction
  • a merge candidate list may be used for coding a block in a merge mode.
  • the motion vector of the block can be represented by one of the candidates MV in the merge MV list. Therefore, instead of transmitting the motion information of the block directly, a merge index is transmitted to a decoder side.
  • the decoder maintains a same merge list and uses the merge index to retrieve the merge candidate as signalled by the merge index.
  • the merge candidate list consists of a small number of candidates and transmitting the merge index is much more efficient than transmitting the motion information.
  • the prediction residuals are still transmitted. In the case that the prediction residuals are zero or very small, the prediction residuals are "skipped" (i.e., the skip mode) and the block is coded by the skip mode with a merge index to identify the merge MV in the merge list.
  • FRUC refers to motion vector derivation for Frame Rate Up-Conversion
  • the underlying techniques are intended for a decoder to derive one or more merge MV candidates without the need for explicitly transmitting motion information. Accordingly, the FRUC is also called decoder derived motion information in this disclosure.
  • the template matching method is a pattern-based MV derivation technique
  • the template matching method of the FRUC is also referred as Pattern-based MV Derivation (PMVD) in this disclosure.
  • PMVD Pattern-based MV Derivation
  • temporal derived MVP is derived by scanning all MVs in all reference pictures.
  • the MV is scaled to point to the current frame.
  • the 4x4 block that pointed by this scaled MV in current frame is the target current block.
  • the MV is further scaled to point to the reference picture that refldx is equal 0 in LIST_0 for the target current block.
  • the further scaled MV is stored in the LIST_0 MV field for the target current block.
  • each small square block corresponds to a 4x4 block.
  • the temporal derived MVPs process scans all the MVs in all 4x4 blocks in all reference pictures to generate the temporal derived LIST_0 and LIST_1 MVPs of current frame.
  • Motion vectors 620 and 630 for two blocks in LIST_0 reference picture with index equal to 1 are known.
  • temporal derived MVP 622 and 632 can be derived by scaling motion vectors 620 and 630 respectively.
  • the scaled MVP is then assigned it to a corresponding block.
  • Motion vectors 650 and 660 for two blocks in LIST_1 reference picture with index equal to 1 are known.
  • temporal derived MVP 652 and 662 can be derived by scaling motion vectors 650 and 660 respectively.
  • the bilateral matching merge mode and template matching merge mode two-stage matching is applied.
  • the first stage is PU-level matching
  • the second stage is the sub-PU-level matching.
  • multiple initial MVs in LIST_0 and LIST_1 are selected respectively.
  • These MVs includes the MVs from merge candidates (i.e., the conventional merge candidates such as these specified in the HEVC standard) and MVs from temporal derived MVPs.
  • Two different starting MV sets are generated for two lists. For each MV in one list, a MV pair is generated by composing of this MV and the mirrored MV that is derived by scaling the MV to the other list.
  • two reference blocks are compensated by using this MV pair. The sum of absolutely differences (SAD) of these two blocks is calculated. The MV pair with the smallest SAD is selected as the best MV pair.
  • SAD absolutely differences
  • HEVC high definition motion model
  • MCP motion compensation prediction
  • x ′ ax + by + e
  • y ′ cx + dy + f .
  • a total of six parameters i.e., a , b, c and ) are used.
  • the motion vector between this pixel and its corresponding reference pixel A'(x', y') can be derived using motion vector field (MVF) of a block.
  • Fig. 7 illustrates an example of motion compensation according to the affine model, where a current area 710 is mapped to a reference area 720 in a reference picture.
  • the affine transform can map any triangle to any triangle.
  • the correspondences between the three corner pixels (i.e., control points) of the current area and the three corner pixels of the reference area can be determined by the three motion vectors (v0, v1 and v2)associated with the three control points as shown in Fig. 7 .
  • the six parameters for the affine model can be derived based on three known motion vectors for three different locations.
  • ( v 0x, v 0y ) is motion vector v 0 of the top-left corner control point
  • ( v 1x , v 1y ) is motion vector v 1 of the top-right corner control point
  • ( v 2x , v 2y ) is motion vector v 2 of the bottom-left corner control point.
  • affine transform motion compensation prediction is applied to improve the coding efficiency.
  • Fig. 8 the affine motion field of the block is described by motion vectors associated with two control points.
  • ( v 0x, v 0y ) is motion vector v 0 of the top-left corner control point and ( v 1x , v 1y ) is motion vector v 1 of the top-right corner control point.
  • block based affine transform prediction is applied.
  • the motion vector of the centre sample of each sub-block is calculated according to the above equation (1), and the result is rounded to 1/16 fraction accuracy.
  • the motion compensation interpolation is applied to generate the prediction of each sub-block with the derived motion vector.
  • the high accuracy motion vector of each sub-block is rounded and saved as the same accuracy as the normal motion vector.
  • a CU/PU is divided into several sub-blocks. Each block can have different reference pictures and different MVs. If the MVs of each sub-block are very diverse, a lot of bandwidth is required for the sub-block motion compensation.
  • OBMC Overlapped block motion compensation
  • the motion vectors associated with the sub-block may be different.
  • the reference pixels of sub-blocks pointed by the sub-block MVs may spread out. This will further increase required system bandwidth. Therefore, it is desired to develop methods to reduce the required bandwidth for sub-block motion compensation coding tools.
  • a MV constraint is applied according to the present invention for the sub-block motion compensation tools, which include ATMVP (Advanced Temporal Motion Vector Prediction), STMVP (Spatial-Temporal Motion Vector Prediction), affine mode, and sub-block refined PMVD (Pattern-based Motion Vector Derivation) mode.
  • the sub-block motion compensation mode is applied to the current block, the MVs of the current block are restricted within a range.
  • the range can be restricted to a range around a primary MV.
  • the MV difference of each sub-block in a CU/PU is restricted with one or more thresholds. For example, a primary MV in a reference list is first derived.
  • All MVs in the reference list in the current block are restricted within the range of primary MV ⁇ a threshold (i.e., from primary MV - a threshold to primary MV + a threshold). If the sub-block MV is outside of the restriction range, the sub-block MV is clipped within the restriction range or replaced by the primary MV. In another example, if MV of one of the list is outside of the restriction range while the MV of the other list is inside the restriction range, only the MV of the other list is used. The MV of the list that is outside of the restriction range is modified as non-valid.
  • the MV is inside or outside the restriction range may be determined according to the reference block pointed by the MV or any other MV related information, which should not be limited in this disclosure.
  • a reference-block range is defined. If the reference block pointed by the MV or any other MV related information is covered by the reference-block range, the MV is considered as inside the restriction range.
  • the threshold can be a predefined value, a derived value, or a signalled value.
  • the threshold can be signalled in the sequence parameter set, picture parameter set, or slice header.
  • the threshold can be a fixed value or an adaptive value.
  • the threshold value can depend on the size of the current block, the width/height of the current block or the current sub-block, the primary MV, the inter prediction direction of the current block or the current sub-block, or the combination of the above.
  • the threshold can be 2 for CU area smaller than 64, and is 16 for CU area equal to or larger than 64.
  • the threshold is larger or un-limited for the uni-predicted block, and is smaller for the bi-predicted block. Multiple thresholds can also be used.
  • one threshold is for the horizontal component of the MV and the other one threshold is for the vertical component of the MV.
  • the horizontal component of the MV threshold can be dependent on the size of the current block or the width and/or height of the current block.
  • the vertical component of the MV threshold can be dependent on the size of the current block or the height and/or width of the current block.
  • the threshold can be different for positive direction and negative direction.
  • the horizontal components of the MVs of a current block are limited within the range of (primary_MVx - thresholdX 1 , primary_MVx + thresholdX 2 ).
  • the thresholdX 1 and thresholdX 2 can depend on the size of the current block, the width and/or height of the current block, or the primary MV.
  • the primary MV can be the default MV, the initial motion vector (vec_init), scaled initial MV (vec_init_scaled), one of the sub-block MVs in the current block (e.g. the MVs of the corner sub-blocks or the MV of the centre sub-blocks), or one derived MV (e.g. the derived MV using the centre pixel or centre block of the current block, or the derived MV using the Sub-PU around the centre position).
  • the initial motion vector vec_init
  • vec_init_scaled the initial motion vector
  • one of the sub-block MVs in the current block e.g. the MVs of the corner sub-blocks or the MV of the centre sub-blocks
  • one derived MV e.g. the derived MV using the centre pixel or centre block of the current block, or the derived MV using the Sub-PU around the centre position.
  • the primary MV can be the first one of at least one derived sub-block MV (e.g. the MV of sub-block A which is the first one derived according to a deriving order in Fig. 3 ), a MV of neighbouring blocks, a temporal collocated MV, or one derived MV.
  • the derived MV can be the MV derived by using the centre pixel or centre block of the current block.
  • the primary MV can be one of the control point MVs, one of the sub-block MVs in current block (e.g. the MVs of the corner sub-blocks or the MV of the centre sub-blocks), or one derived MV (e.g. the derived MV using the centre pixel or centre block of the current block).
  • the MVs of the control points are outside of the restriction range, the MVs of the control points are first clipped within the restriction range or replaced by the primary MV. After the clipping or replacement, the MVs of each sub-block are derived by using the new control point MVs.
  • the primary MV can be the PU initial MV or the CU/PU refined MV.
  • the sub-block refined MV should be limited around the primary MV.
  • the MV threshold can be different for different sub-blocks.
  • the horizontal MV of the sub-block 1 is within the range of (primary_MVx - thresholdX 1 , primary_MVx + thresholdX 2 ), and the horizontal MV of the sub-block 2 is within the range of (primary_MVx - thresholdX 3 , primary _MVx + thresholdX 4 ).
  • the thresholdX 1 can be different from thresholdX 3 . For example, if the location of the sub-block 1 is left to the location of the sub-block 2, the thresholdX 1 can be smaller than thresholdX 3 .
  • the proposed method can be applied normatively or can be applied as encoder-side constraint. If it's normative, the MVs of all sub-block are generated with the rule as described above. All the MVs will be within the restriction range around the primary MV. For encoder constraint, it is a requirement of bitstream conformance that the MVs of all sub-blocks shall follow the rule as describe before. All the MVs of the sub-blocks shall be within the restriction range around the primary MV. If not, this mode or this merge candidate shall not be selected in the encoder side.
  • the primary MV is first derived.
  • the decoder can load the reference data around the primary MV. For example, the block with the size of (thresholdX1+blockWidth+interpolationFilterTapLength-1+thresholdX2) * (thresholdX1+blockHeight+ interpolationFilterTapLength -1+thresholdX2) can be loaded.
  • the sub-block MV is not limited to be around the primary MV.
  • a reference block (referred as primary reference block in this disclosure) is used for the motion compensation of all sub-blocks.
  • the primary reference block covers a block pointed by the primary MV.
  • pseudo reference pixels are used.
  • the pseudo reference pixels can be a predefined pixel value or the padding pixels of the reference block.
  • the size and shape of the reference block can depend on the current CU size and shape. Fig. 9 shows an example.
  • a primary MV 922 is first derived.
  • Primary reference block 920 pointed by the primary MV 922 in reference picture 930 is used for the motion compensation of the sub-blocks of the 8x8 CU. If all the reference pixels of a sub-blocks are within the primary reference block 920(e.g. block A, all the reference pixels of which are within the primary reference block 920), the pixels in the reference block are used. If the reference pixels of a sub-blocks are not within the primary reference block 920(e.g. block C, reference pixels of which are not within the primary reference block 920), the padded pixels according to the reference block are used.
  • part of the reference pixels of a sub-blocks are within the primary reference block and part of the reference pixels are not (e.g. block B, which has part of the reference pixels within the primary reference block 920 and has part of the reference pixels not within the primary reference block 920), part of pixels in the primary reference block and part of padded pixels are used.
  • the interpolation filter coefficients are modified to exclude any reference pixel of reference sub-blocks outside the primary reference block.
  • Fig. 10 illustrates an exemplary flowchart of a video coding system using constrained sub-block motion vectors according to a non-claimed example.
  • the steps shown in the flowchart may be implemented as program codes executable on one or more processors (e.g., one or more CPUs) at the encoder side.
  • the steps shown in the flowchart may also be implemented based hardware such as one or more electronic devices or processors arranged to perform the steps in the flowchart.
  • input data associated with a current block in a current picture is received in step 1010.
  • the input data may correspond to pixel data to be encoded at a video encoder side or a video bitstream corresponding to compressed data including the current block at a video decoder side.
  • step 1020 Whether a sub-block motion compensation coding tool is selected for the current block is checked in step 1020. If a sub-block motion compensation coding tool is selected for the current block (i.e., the "yes" path from step 1020), steps 1040 to 1060 are performed. Otherwise (i.e., the "no" path from step 1020), steps 1040 to 1060 are skipped.
  • step 1040 sub-block MVs (motion vectors) associated with multiple sub-blocks are generated according to the sub-block motion compensation coding tool, wherein the current block includes or contains the multiple sub-blocks.
  • step 1050 the sub-block MVs are constrained within a range to form constrained sub-block MVs.
  • step 1060 motion compensation is applied to the current block using the constrained sub-block MVs or motion compensation is applied to the current block using one sub-block MV within the range in a second list if a corresponding sub-block MV in a first list is outside the range.
  • Fig. 11 illustrates another exemplary flowchart of a video coding system using constrained sub-block motion vectors according to a non-claimed example.
  • input data associated with a current block in a current picture is received in step 1110.
  • the input data may correspond to pixel data to be encoded at a video encoder side or a video bitstream corresponding to compressed data including the current block at a video decoder side.
  • Whether a sub-block motion compensation coding tool is selected for the current block is checked in step 1120. If a sub-block motion compensation coding tool is selected for the current block (i.e., the "yes" path from step 1120), steps 1130 to 1160 are performed.
  • steps 1130 to 1160 are skipped.
  • a primary reference block is derived.
  • sub-block MVs motion vectors associated with multiple sub-blocks are generated according to the sub-block motion compensation coding tool, wherein the current block includes or contains the multiple sub-blocks.
  • motion compensation is applied to the current block using one or more reference pixels of reference sub-blocks within the primary reference block and excluding any reference pixel of reference sub-blocks outside the primary reference block, wherein the reference sub-blocks are pointed by sub-block MVs.
  • Fig. 12 illustrates an example of bandwidth analysis for constrained MVs according to a non-claimed example.
  • OBMC is enabled and the current block is a bi-prediction block.
  • the current block 1210 is 8x8, which includes or contains four 4x4 sub-blocks for sub-block based motion compensation.
  • Fig. 12 illustrates the required reference pixels 1220 to be accessed in order to perform the sub-block based motion compensation. For example, the additional pixel data outside the collocated block are indicated for the upper-left sub-block (labelled as UL in Fig. 12 ) and the lower-right sub-block (labelled as LR in Fig. 12 ).
  • UL upper-left sub-block
  • LR lower-right sub-block
  • LF refers to the required number of pixels on the left side of a sub-block for interpolation filter
  • RF refers to the required number of pixels on the right side of a sub-block for interpolation filter
  • OBMC refers to the required number of pixels for OBMC on one boundary.
  • OBMC refers to the required number of pixels for OBMC on one boundary.
  • the threshold will be 3.5 according to a non-claimed example of the present invention since 19 pixels (i.e., 3+8+4+4) are already requires and only 3.5 pixels are remaining (i.e., (26-19)/2).
  • the threshold for other coding conditions can be derived similarly.
  • any of the foregoing proposed methods can be implemented in encoders and/or decoders.
  • any of the proposed methods can be implemented in a sub-block partitioning module or a MV derivation module of an encoder, and/or a sub-block partitioning module or a MV derivation module of a decoder.
  • any of the proposed methods can be implemented as a circuit coupled to the sub-block partitioning module or the MV derivation module of the encoder and/or the sub-block partitioning module or the MV derivation module of the decoder, so as to provide the information needed by the sub-block partitioning module or the MV derivation module.
  • Embodiment of the present invention as described above may be implemented in various hardware, software codes, or a combination of both.
  • an embodiment of the present invention can be one or more circuit circuits integrated into a video compression chip or program code integrated into video compression software to perform the processing described herein.
  • An embodiment of the present invention may also be program code to be executed on a Digital Signal Processor (DSP) to perform the processing described herein.
  • DSP Digital Signal Processor
  • the invention may also involve a number of functions to be performed by a computer processor, a digital signal processor, a microprocessor, or field programmable gate array (FPGA). These processors can be configured to perform particular tasks according to the invention, by executing machine-readable software code or firmware code that defines the particular methods embodied by the invention.
  • the software code or firmware code may be developed in different programming languages and different formats or styles.
  • the software code may also be compiled for different target platforms.
  • different code formats, styles and languages of software codes and other means of configuring code to perform the tasks in accordance with the invention will not depart from the scope of the invention.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Compression Or Coding Systems Of Tv Signals (AREA)

Claims (7)

  1. Ein Verfahren (1010, 1020, 1040, 1050, 1060) zur Videokodierung unter Verwendung eines Inter-Prediction-Modus in einem Videokodiersystem, wobei das Verfahren Folgendes umfasst:
    Empfangen (1010) von Eingabedaten, die einem aktuellen Block in einem aktuellen Bild zugeordnet sind;
    wenn für den aktuellen Block ein Subblock-Bewegungskompensations-Kodierungstool ausgewählt wird (1020):
    Erzeugen (1040) von Subblock-Bewegungsvektoren (MVs), die mehreren Subblöcken gemäß dem Subblock-Bewegungskompensations-Kodierungstool zugeordnet sind, wobei der aktuelle Block die mehreren Subblöcke umfasst;
    Einschränken (1050) jedes der Subblock-Bewegungsvektoren innerhalb eines Bereichs eines primären Bewegungsvektors ± einem Schwellenwert, um eingeschränkte Subblock-Bewegungsvektoren zu bilden; und
    Anwenden (1060) einer Bewegungskompensation auf den aktuellen Block unter Verwendung der eingeschränkten Subblock-Bewegungsvektoren,
    wobei das Subblock-Bewegungskompensations-Kodierungstool aus einer Gruppe ausgewählt ist, die den Modus "Advanced Temporal Motion Vector Prediction", ATMVP, "Spatial-Temporal Motion Vector Prediction", STMVP, affine Vorhersage und "Subblock Refined Pattern-based Motion Vector Derivation" (PMVD) umfasst, wobei der Modus der Subblock-verfeinerten musterbasierten Bewegungsvektorableitung (PMVD) bilaterales Matching oder zeitliches Matching verwendet,
    wobei, wenn das Subblock-Bewegungskompensations-Kodierungstool das ATMVP ist, der primäre Bewegungsvektor ein Standard-Bewegungsvektor, ein anfänglicher Bewegungsvektor, ein skalierter anfänglicher Bewegungsvektor, einer der Subblock-Bewegungsvektoren im aktuellen Block oder ein abgeleiteter Bewegungsvektor ist, wobei der abgeleitete Bewegungsvektor ein Bewegungsvektor ist, der unter Verwendung eines Mittelpixels oder eines Mittelblocks des aktuellen Blocks oder unter Verwendung einer Sub-Prediction Unit, Sub-PU, um eine Mittelposition herum abgeleitet wird; wenn das Subblock-Bewegungskompensations-Kodierungstool das STMVP ist, der primäre Bewegungsvektor ein erster abgeleiteter Subblock-Bewegungsvektor, ein Bewegungsvektor benachbarter Blöcke, ein zeitlich kollozierter Bewegungsvektor oder ein abgeleiteter Bewegungsvektor ist, wobei der abgeleitete Bewegungsvektor ein Bewegungsvektor ist, der durch Verwendung eines mittleren Pixel oder eines mittleren Blocks des aktuellen Blocks abgeleitet wird; wenn das Subblock-Bewegungskompensations-Kodierungstool die affine Vorhersage ist, der primäre Bewegungsvektor einer der Kontrollpunkt-Bewegungsvektoren, einer der Subblock-Bewegungsvektoren im aktuellen Block oder ein abgeleiteter Bewegungsvektor ist, wobei der abgeleitete Bewegungsvektor ein unter Verwendung eines zentralen Pixels oder eines zentralen Blocks des aktuellen Blocks abgeleiteter Bewegungsvektor ist;
    wenn das Subblock-Bewegungskompensations-Kodierungstool der PMVD-Modus (Subblock Refined Pattern-Based Motion Vector Derivation) ist, ist der primäre Bewegungsvektor ein PU-Anfangs-Bewegungsvektor oder ein CU/PU-verfeinerter Bewegungsvektor,
    wobei der Schwellenwert für verschiedene Subblöcke unterschiedlich ist, wobei ein horizontaler Bewegungsvektor eines ersten Subblocks im Bereich von (primärer_MVx - SchwellenwertX1, primärer_MVx + SchwellenwertX2) liegt und ein horizontaler Bewegungsvektor eines zweiten Subblocks im Bereich von (primärer_MVx - SchwellenwertX3, primärer_MVx + SchwellenwertX4) liegt, wobei der SchwellenwertX1 sich vom SchwellenwertX3 unterscheidet, wobei falls die Position des ersten Subblocks der Position des zweiten Subblocks überlassen wird, der SchwellenwertX1 kleiner als der SchwellenwertX3 ist.
  2. Das Verfahren (1010, 1020, 1040, 1050, 1060) nach Anspruch 1, wobei der Schwellenwert einem Schwellenwert entspricht und wenn ein Subblock-Bewegungsvektor außerhalb des Bereichs liegt, der Subblock-Bewegungsvektor auf den Bereich begrenzt wird.
  3. Das Verfahren (1010, 1020, 1040, 1050, 1060) nach Anspruch 1, wobei jeder der einen oder mehreren Schwellenwerte einem vordefinierten Wert, einem abgeleiteten Wert oder einem signalisierten Wert entspricht.
  4. Das Verfahren (1010, 1020, 1040, 1050, 1060) nach Anspruch 1, wobei der eine oder die mehreren Schwellenwerte in Abhängigkeit von einer Größe, Breite oder Höhe des aktuellen Blocks oder aktuellen Subblocks, dem primären Bewegungsvektor, der Interprädiktionsrichtung des aktuellen Blocks oder des aktuellen Subblocks oder die Kombination der oben genannten bestimmt werden.
  5. Das Verfahren (1010, 1020, 1040, 1050, 1060) nach Anspruch 1, wobei, wenn das Subblock-Bewegungskompensations-Kodierungstool für den aktuellen Block an einem Videokodierer ausgewählt wird, der Videokodierer die Subblock-Bewegungsvektoren einschränkt, so dass sie innerhalb des Bereichs liegen, um eingeschränkte Subblock-Bewegungsvektoren zu bilden, und Informationen signalisiert, die sich auf die eingeschränkten Subblock-Bewegungsvektoren in einem Videobitstrom beziehen.
  6. Das Verfahren (1010, 1020, 1040, 1050, 1060) nach Anspruch 1, wobei, wenn das Subblock-Bewegungskompensations-Kodierungstool für den aktuellen Block an einem Videocodierer ausgewählt wird, alle Subblock-Bewegungsvektoren beschränkt sind, so dass sie innerhalb eines Bereichs um einen primären Bewegungsvektor liegen und Informationen signalisiert, die sich auf die Subblock-Bewegungsvektoren in einem Videobitstrom beziehen.
  7. Eine Vorrichtung zur Videokodierung unter Verwendung eines Inter-Prediction-Modus in einem Videokodiersystem, wobei die Vorrichtung ein oder mehrere elektronische Geräte oder Prozessoren umfasst, die konfiguriert sind zum:
    Empfangen (1010) von Eingabedaten, die einem aktuellen Block in einem aktuellen Bild zugeordnet sind;
    wenn für den aktuellen Block ein Subblock-Bewegungskompensations-Kodierungstool ausgewählt wird (1020):
    Erzeugen (1040) von Subblock-Bewegungsvektoren (MVs), die mehreren Subblöcken gemäß dem Subblock-Bewegungskompensations-Kodierungstool zugeordnet sind, wobei der aktuelle Block die mehreren Subblöcke umfasst;
    Einschränken (1050) jedes der Subblock-Bewegungsvektoren innerhalb eines Bereichs eines primären Bewegungsvektors ± einem Schwellenwert, um eingeschränkte Subblock-Bewegungsvektoren zu bilden; und
    Anwenden (1060) einer Bewegungskompensation auf den aktuellen Block unter Verwendung der eingeschränkten Subblock-Bewegungsvektoren,
    wobei das Subblock-Bewegungskompensations-Kodierungstool aus einer Gruppe ausgewählt ist, die den Modus "Advanced Temporal Motion Vector Prediction", ATMVP, "Spatial-Temporal Motion Vector Prediction", STMVP, affine Vorhersage und "Subblock Refined Pattern-based Motion Vector Derivation" (PMVD) umfasst, wobei der Modus der Subblock-verfeinerten musterbasierten Bewegungsvektorableitung (PMVD) bilaterales Matching oder zeitliches Matching verwendet,
    wobei, wenn das Subblock-Bewegungskompensations-Kodierungstool das ATMVP ist, der primäre Bewegungsvektor ein Standard-Bewegungsvektor, ein anfänglicher Bewegungsvektor, ein skalierter anfänglicher Bewegungsvektor, einer der Subblock-Bewegungsvektoren im aktuellen Block oder ein abgeleiteter Bewegungsvektor ist, wobei der abgeleitete Bewegungsvektor ein Bewegungsvektor ist, der unter Verwendung eines Mittelpixels oder eines Mittelblocks des aktuellen Blocks oder unter Verwendung einer Sub-Prediction Unit, Sub-PU, um eine Mittelposition herum abgeleitet wird; wenn das Subblock-Bewegungskompensations-Kodierungstool das STMVP ist, der primäre Bewegungsvektor ein erster abgeleiteter Subblock-Bewegungsvektor, ein Bewegungsvektor benachbarter Blöcke, ein zeitlich kollozierter Bewegungsvektor oder ein abgeleiteter Bewegungsvektor ist, wobei der abgeleitete Bewegungsvektor ein Bewegungsvektor ist, der durch Verwendung eines mittleren Pixel oder eines mittleren Blocks des aktuellen Blocks abgeleitet wird; wenn das Subblock-Bewegungskompensations-Kodierungstool die affine Vorhersage ist, der primäre Bewegungsvektor einer der Kontrollpunkt-Bewegungsvektoren, einer der Subblock-Bewegungsvektoren im aktuellen Block oder ein abgeleiteter Bewegungsvektor ist, wobei der abgeleitete Bewegungsvektor ein unter Verwendung eines zentralen Pixels oder eines zentralen Blocks des aktuellen Blocks abgeleiteter Bewegungsvektor ist;
    wenn das Subblock-Bewegungskompensations-Kodierungstool der PMVD-Modus (Subblock Refined Pattern-Based Motion Vector Derivation) ist, ist der primäre Bewegungsvektor ein PU-Anfangs-Bewegungsvektor oder ein CU/PU-verfeinerter Bewegungsvektor,
    wobei der Schwellenwert für verschiedene Subblöcke unterschiedlich ist, wobei ein horizontaler Bewegungsvektor eines ersten Subblocks im Bereich von (primärer_MVx - SchwellenwertX1, primärer_MVx + SchwellenwertX2) liegt und ein horizontaler Bewegungsvektor eines zweiten Subblocks im Bereich von (primärer_MVx - SchwellenwertX3, primärer_MVx + SchwellenwertX4) liegt, wobei der SchwellenwertX1 sich vom SchwellenwertX3 unterscheidet, wobei falls die Position des ersten Subblocks der Position des zweiten Subblocks überlassen wird, der SchwellenwertX1 kleiner als der SchwellenwertX3 ist.
EP18801342.9A 2017-05-18 2018-05-17 Verfahren und vorrichtung zur bewegungsvektorbeschränkung für videocodierung Active EP3616406B1 (de)

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
US201762507838P 2017-05-18 2017-05-18
US201762519214P 2017-06-14 2017-06-14
PCT/CN2018/087351 WO2018210315A1 (en) 2017-05-18 2018-05-17 Method and apparatus of motion vector constraint for video coding

Publications (3)

Publication Number Publication Date
EP3616406A1 EP3616406A1 (de) 2020-03-04
EP3616406A4 EP3616406A4 (de) 2020-11-11
EP3616406B1 true EP3616406B1 (de) 2024-03-20

Family

ID=64273342

Family Applications (1)

Application Number Title Priority Date Filing Date
EP18801342.9A Active EP3616406B1 (de) 2017-05-18 2018-05-17 Verfahren und vorrichtung zur bewegungsvektorbeschränkung für videocodierung

Country Status (6)

Country Link
US (2) US11700391B2 (de)
EP (1) EP3616406B1 (de)
CN (1) CN110651474B (de)
CA (1) CA3063746A1 (de)
TW (1) TWI687096B (de)
WO (1) WO2018210315A1 (de)

Families Citing this family (20)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9510009B2 (en) * 2010-05-20 2016-11-29 Thomson Licensing Methods and apparatus for adaptive motion vector candidate ordering for video encoding and decoding
CN109005412B (zh) * 2017-06-06 2022-06-07 北京三星通信技术研究有限公司 运动矢量获取的方法及设备
KR20190108506A (ko) * 2018-03-14 2019-09-24 한국전자통신연구원 영상 부호화/복호화 방법, 장치 및 비트스트림을 저장한 기록 매체
CN118075486A (zh) * 2018-06-04 2024-05-24 华为技术有限公司 获取运动矢量的方法和装置
WO2020084475A1 (en) 2018-10-22 2020-04-30 Beijing Bytedance Network Technology Co., Ltd. Utilization of refined motion vector
CN117459722A (zh) 2018-11-12 2024-01-26 北京字节跳动网络技术有限公司 组合帧间-帧内预测的简化
EP3861742A4 (de) 2018-11-20 2022-04-13 Beijing Bytedance Network Technology Co., Ltd. Differenzberechnung basierend auf einer räumlichen position
KR20240011246A (ko) * 2018-11-26 2024-01-25 프라운호퍼 게젤샤프트 쭈르 푀르데룽 데어 안겐반텐 포르슝 에. 베. 타일 독립성 제약을 이용하는 인터 예측 개념
WO2020108650A1 (en) * 2018-11-29 2020-06-04 Beijing Bytedance Network Technology Co., Ltd. Flag indication method in intra block copy mode
WO2020114515A1 (en) * 2018-12-08 2020-06-11 Beijing Bytedance Network Technology Co., Ltd. Reducing the in-ctu storage required by affine inheritance
CN111316646B (zh) * 2019-01-02 2023-06-23 北京大学 候选mv列表构建方法和装置
CN113302938A (zh) * 2019-01-11 2021-08-24 北京字节跳动网络技术有限公司 整数mv运动补偿
CN113302918A (zh) * 2019-01-15 2021-08-24 北京字节跳动网络技术有限公司 视频编解码中的加权预测
WO2020177755A1 (en) 2019-03-06 2020-09-10 Beijing Bytedance Network Technology Co., Ltd. Usage of converted uni-prediction candidate
JP2020150294A (ja) * 2019-03-11 2020-09-17 ルネサスエレクトロニクス株式会社 復号方法、復号装置及び符号化方法
CN111698515B (zh) * 2019-03-14 2023-02-14 华为技术有限公司 帧间预测的方法及相关装置
US11343525B2 (en) 2019-03-19 2022-05-24 Tencent America LLC Method and apparatus for video coding by constraining sub-block motion vectors and determining adjustment values based on constrained sub-block motion vectors
WO2020192747A1 (en) * 2019-03-27 2020-10-01 Beijing Bytedance Network Technology Co., Ltd. Motion information precision alignment in affine advanced motion vector prediction
CN114930840A (zh) * 2020-01-07 2022-08-19 华为技术有限公司 增强型插值滤波器的运动矢量范围的推导
US20230224472A1 (en) * 2022-01-12 2023-07-13 Tencent America LLC Motion vector restriction for out-of-frame boundary conditions

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP3766247A1 (de) * 2018-04-02 2021-01-20 MediaTek Inc. Videoverarbeitungsverfahren und -vorrichtungen zur subblockbewegungskompensation bei videocodiersystemen

Family Cites Families (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100098169A1 (en) * 2008-10-16 2010-04-22 Texas Instruments Incorporated Method and apparatus for motion estimation using compressed reference frame
KR20120088488A (ko) * 2011-01-31 2012-08-08 한국전자통신연구원 시간적 움직임 벡터 저장 방법 및 그 장치
KR102211673B1 (ko) 2011-12-16 2021-02-03 벨로스 미디어 인터내셔널 리미티드 동화상 부호화 방법, 동화상 부호화 장치, 동화상 복호 방법, 동화상 복호 장치, 및 동화상 부호화 복호장치
US9762904B2 (en) 2011-12-22 2017-09-12 Qualcomm Incorporated Performing motion vector prediction for video coding
CN108377393A (zh) 2012-03-22 2018-08-07 联发科技股份有限公司 编码视频数据的方法以及装置
US10021414B2 (en) 2013-01-04 2018-07-10 Qualcomm Incorporated Bitstream constraints and motion vector restriction for inter-view or inter-layer reference pictures
CN105393539B (zh) 2013-07-24 2019-03-29 高通股份有限公司 用于纹理及深度译码的子pu运动预测
WO2015169230A1 (en) * 2014-05-06 2015-11-12 Mediatek Inc. Video processing method for determining position of reference block of resized reference frame and related video processing apparatus
US11140400B2 (en) * 2015-06-05 2021-10-05 Sony Corporation Image processing apparatus and image processing method for intra BC prediction
WO2017036414A1 (en) * 2015-09-02 2017-03-09 Mediatek Inc. Method and apparatus of decoder side motion derivation for video coding
US10448010B2 (en) * 2016-10-05 2019-10-15 Qualcomm Incorporated Motion vector prediction for affine motion models in video coding
US11277635B2 (en) * 2017-03-17 2022-03-15 Vid Scale, Inc. Predictive coding for 360-degree video based on geometry padding
US10382516B2 (en) * 2017-05-09 2019-08-13 Apple Inc. Detecting upscaled source video

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP3766247A1 (de) * 2018-04-02 2021-01-20 MediaTek Inc. Videoverarbeitungsverfahren und -vorrichtungen zur subblockbewegungskompensation bei videocodiersystemen

Also Published As

Publication number Publication date
EP3616406A1 (de) 2020-03-04
EP3616406A4 (de) 2020-11-11
TW201902222A (zh) 2019-01-01
US11700391B2 (en) 2023-07-11
TWI687096B (zh) 2020-03-01
WO2018210315A1 (en) 2018-11-22
US20230300370A1 (en) 2023-09-21
CN110651474B (zh) 2022-02-18
US20210176485A1 (en) 2021-06-10
CA3063746A1 (en) 2018-11-22
CN110651474A (zh) 2020-01-03

Similar Documents

Publication Publication Date Title
EP3616406B1 (de) Verfahren und vorrichtung zur bewegungsvektorbeschränkung für videocodierung
US11956462B2 (en) Video processing methods and apparatuses for sub-block motion compensation in video coding systems
US11895291B2 (en) Method and apparatus of video coding using bi-directional CU weight
US10750203B2 (en) Method and apparatus of adaptive bi-prediction for video coding
WO2017118409A1 (en) Method and apparatus for affine merge mode prediction for video coding system
WO2020015706A1 (en) Method and apparatus of motion compensation bandwidth reduction for video coding system utilizing multi-hypothesis
TW201944781A (zh) 視訊編解碼系統中具有重疊塊運動補償的視訊處理的方法以及裝置
US11856194B2 (en) Method and apparatus of simplified triangle merge mode candidate list derivation
US11539940B2 (en) Method and apparatus of multi-hypothesis in video coding
US11539977B2 (en) Method and apparatus of merge with motion vector difference for video coding
WO2019223790A1 (en) Method and apparatus of affine mode motion-vector prediction derivation for video coding system
KR102463478B1 (ko) 비디오 코딩 시스템을 위한 아핀 인터 예측 방법 및 장치
WO2024078331A1 (en) Method and apparatus of subblock-based motion vector prediction with reordering and refinement in video coding
WO2024016844A1 (en) Method and apparatus using affine motion estimation with control-point motion vector refinement
WO2024027784A1 (en) Method and apparatus of subblock-based temporal motion vector prediction with reordering and refinement in video coding

Legal Events

Date Code Title Description
STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE INTERNATIONAL PUBLICATION HAS BEEN MADE

PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE

17P Request for examination filed

Effective date: 20191127

AK Designated contracting states

Kind code of ref document: A1

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

AX Request for extension of the european patent

Extension state: BA ME

DAV Request for validation of the european patent (deleted)
DAX Request for extension of the european patent (deleted)
A4 Supplementary search report drawn up and despatched

Effective date: 20201013

RIC1 Information provided on ipc code assigned before grant

Ipc: H04N 19/503 20140101AFI20201007BHEP

Ipc: H04N 19/55 20140101ALI20201007BHEP

Ipc: H04N 19/513 20140101ALI20201007BHEP

RAP1 Party data changed (applicant data changed or rights of an application transferred)

Owner name: HFI INNOVATION INC.

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: EXAMINATION IS IN PROGRESS

17Q First examination report despatched

Effective date: 20220524

P01 Opt-out of the competence of the unified patent court (upc) registered

Effective date: 20230627

GRAP Despatch of communication of intention to grant a patent

Free format text: ORIGINAL CODE: EPIDOSNIGR1

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: GRANT OF PATENT IS INTENDED

INTG Intention to grant announced

Effective date: 20231009

GRAS Grant fee paid

Free format text: ORIGINAL CODE: EPIDOSNIGR3

GRAA (expected) grant

Free format text: ORIGINAL CODE: 0009210

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE PATENT HAS BEEN GRANTED

AK Designated contracting states

Kind code of ref document: B1

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

REG Reference to a national code

Ref country code: GB

Ref legal event code: FG4D

REG Reference to a national code

Ref country code: CH

Ref legal event code: EP

REG Reference to a national code

Ref country code: IE

Ref legal event code: FG4D

REG Reference to a national code

Ref country code: DE

Ref legal event code: R096

Ref document number: 602018066917

Country of ref document: DE